Pillar
Internal tooling for a 400-person ops team

Pillar's operations team had grown from 40 to 400 people in three years. The tools hadn't kept up. Fourteen spreadsheets tracked different workflows, three SaaS products covered overlapping territory with no integration between them, and the ops managers spent more time maintaining the tooling than doing the work the tools were supposed to support.

We started by shadowing five ops managers for a full week. Not interviewing — watching. The gap between what they said their workflow was and what it actually was turned out to be the most valuable input of the entire engagement. Three of the fourteen spreadsheets existed solely to reconcile data between two SaaS tools that should have been talking to each other.

The platform we built is a Next.js frontend with a Laravel API (Pillar's existing backend was Laravel and their team knew how to maintain it) on Postgres. We designed the workspace around the actual workflows we'd observed: task queues that matched how ops managers actually triaged work, dashboards that answered the three questions they checked every morning, and an audit trail that eliminated the 'who changed this' conversations that ate hours every week.

The 9,200-hour estimate came from Pillar's ops leadership six months after launch. It's a rough number — they calculated it by comparing the time their team spent on tooling maintenance before and after. The real number might be higher, but they were conservative and so are we.